View
6
Download
0
Category
Preview:
Citation preview
Blood, Vol. 58, No. 1 (July), 1981
BLOOD The American
The Journal of
Society of Hematology
VOL. 58, NO. 1 JULY 1981
REVIEW
The Factor VIII Complex: Structure and Function
By Leon W. Hoyer
Normal human plasma contains a complex of two proteins
that are important in hemostasis and coagulation. The
factor VIII procoagulant protein (antihemophilic factor) and
the factor VIll-related protein (von Willebrand factor) are
under separate genetic control, have distinct biochemical
and immunologic properties, and have unique and essential
T HE IMPORTANCE of factor VIII in hemostasis
and blood coagulation is obvious from the clini-
cal problems in the factor VIII deficiency diseases,
classic hemophilia, and von Willebrand’s disease.
During the past decade there has been intense
research interest in these diseases, the two most
common hereditary bleeding disorders, and in the
properties of factor VIII. These studies have led to an
evolving understanding of factor VIII structure and
function.
The concept that factor VIII has two distinct
biologic functions-coagulant activity and a role in
primary hemostasis-was first suggested as an expla-
nation for the dual defect in von Willebrand’s disease.’
The logical inference that factor VIII is a bifunctional
molecule was strengthened by reports that proteins
purified from human and bovine plasmas had both
factor VIII procoagulant activity and the capacity to
interact with platelets in a way that might reflect an in
vivo role in primary hemostasis.25 Subsequent studies
have suggested an alternative interpretation, and it is
now generally accepted that plasma factor VIII is a
complex of two components that have distinct func-
tions, biochemical and immunologic properties, and
genetic control. The properties of these components
are summarized in Table I and the relationship is
illustrated in Fig. I.
One component of the factor VIII complex has
antihemophilic factor procoagulant activity and is now
usually designated VlII:C. It is inactivated by human
antibodies and can be measured (as VllI:CAg) when
these reagents are used for immunoassays. The other,
physiologic functions. While the nature of their interaction
and the details of the biochemical structures remain to be
determined, the information now available permits a
preliminary understanding of the molecular defects in
hemophilia and von Willebrand’s disease.
larger component comprises the majority of the
protein mass, interacts with platelets in a way that
promotes primary hemostasis, and can be immunopre-
cipitated by heterologous antisera. It is usually desig-
nated factor Vill-related protein (VIIIR) or von
Willebrand factor since it is reduced in quantity or is
qualitatively abnormal in von Willebrand’s disease.
Although it has been suggested that the two compo-
nents are properties of a single macromolecule,6’7
several kinds of data demonstrate the essential differ-
ences of the two proteins.
(1) The factor VIII procoagulant protein and the
factor Vill-related protein are controlled by different
genes. Isolated VlII:C deficiency is characteristic of
hemophilia, a disease transmitted by X-chromosomal
inheritance. In contrast, reduced or abnormal VIIIR is
found in von Willebrand’s disease, and the inheritance
pattern is that of an autosomal gene.
(2) The two proteins can be separated by chroma-
tography or centrifugation in high ionic strength
buffers (1 M NaCI or 0.24 M CaCI2).8” In most
studies, the inclusion of protease inhibitors in the
buffers does not affect the separation.’2”3
From the Hematology Division, Department of Medicine,
University of Connecticut School of Medicine, Farmington, Conn.
Supported in part by USPHS Research Grants HL 16626 and
HL 16872.
Submitted November 19, 1980, accepted February 12, 1981.
Address reprint requests to Leon W. Hoyer, M.D., University of
Connecticut Health Center, Farmington, Conn. 06032.
(C I 981 by Grune & Stratton, Inc.
0006-4971/81/5801-000l$02.00/0
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
- ? CELL - ENDOTHELIAL CELL
- MEGAKARYOCYTE
�ThVIII:C VIIIR SUBUNIT
/VIIIR POLYMER
(VIII:C) (VIIIRPOLYMER)
VIII:C
+
VI IIRPOLYMER
HIGH IONICSTRENGTH REDUCTION VIII:C
+
2 LEON W. HOVER
THE FACTOR VIII COI4�LEX IN PLASMA
Table 1 . The Components of the Factor VIII Complex
Vlll:C The factor VIII procoagulant protein: the antihemo-
philic factor.
Identified as:
Factor VIII procoagulant activity (VlIl:C)
The procoagulant property of normal plasma
that is measured in standard coagulation
assays.
Factor VIII procoagulant antigen (Vlll:CAg)
Antigenic determinants closely associated
with VIII:C. Measured by immunoassays
carried out with human antibodies.
VIIIR The factor VIII-related protein: the von Willebrand factor.
A large polymeric protein that is necessary
for normal platelet adhesion and bleeding
time in vivo.
Identified as:
Factor-VllI-related antigen (VIIIR:Ag)
Antigenic determinants on VIIIR that are
detected by heterologous antibodies.
Ristocetin cofactor (VIIIR:RC)
The property of normal plasma VIIIR that
supports ristocetin-induced agglutination of
washed normal platelets.
(3) The plasma concentration of the two proteins,
measured by both function and immunoassays, vary
independently under certain conditions, the most strik-
ing being the posttransfusion period in patients with
von Willebrand’s disease.’
(4) The two proteins have different antigenic deter-
minants. Factor VIII procoagulant activity is macti-
vated by human antibodies from multitransfused
hemophiliacs and patients with spontaneous inhibitors;
ristocetin cofactor activity (VIIIR:RC) and the bleed-
ing time are characteristically normal in these
patients. Each protein can be measured by an immu-
noassay that is independent of the other protein.’4”5
Human anti-VIII:C coupled to Sepharose remove
x�-GIRc11L&PE
VIIl:C (and VIII:CAg) from plasma, but there is no
change in VIIIR:Ag levels.’6
(5) The biologic properties of the two proteins are
also independent. The procoagulant protein retains
full VIII:C activity in the virtual absence of VIIIR
(i.e., an VIII:C/VIIIR:Ag ratio of 1 2,600: 1 ),17 and the
factor-Vill-related protein can have VIIIR:RC activ-
ity in the absence ofdetectable VIII:C.”
The two components of the factor VIII complex do
interact, however. Their concentrations vary together
under most normal, stressful, and pathologic situa-
tions,’9 and standard purification methods separate
the intact (two component) factor VIII complex from
other plasma proteins. This interaction will be consid-
ered in detail after a description of the properties of
the two components.
FACTOR VIII PROCOAGULANT PROTEIN:
ANTIHEMOPHILIC FACTOR
Biochemical Properties
At the present time, there is little published infor-
mation about the biochemical properties of the factor
VIII procoagulant protein, per se. With few excep-
tions, studies of VIII:C function have been carried out
with the intact factor VIII complex, and it is only
recently that the factor VIII procoagulant protein has
been characterized after separation from VIIIR as
well as from the other plasma pr���jfl5#{149}l72O2�
Bovine VIII:C has been purified approximately
300,000-fold from plasma by Vehar and Davie.2#{176}
Although the G-200 gel filtration properties of the
protein with VIII:C activity suggested a molecular
weight of 250,000-300,000, analysis in sodium dode-
cyl sulfate/urea-polyacrylamide gel electrophoresis
identified a triplet of protein-staining bands with
Fig. 1 . The factor VIII com-plex. This schematic interpreta-tion indicates the interaction of
the two components and theirgenetic control. The effects of
VIIIR high ionic strength and reductionSUBUNITS are also noted.
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
AControl
Vo 0 A
A
ci,
6 .20’
�?-�=�
h .20’
U��%
>�-‘�
? .‘� .10�IC
IDU
5 0�
cct .10
c�E�
4
�o , ,
60 80 100 120
Elution Volume (ml)
Fig. 2. The Sephadex G-200 gel filtration properties of VIII:C
that has been separated from VIIIR. Vlll:C procoagulant andVlll:CAg measurements are indicated. For reference, the void
volume (V0) is noted, as are the elution volumes of lgG (G) andalbumin (A). (Modified from Hoyer and Trabold.2’ ) (A) VIII:C free ofdetectable VIllA. (B) Thrombin-activated Vlll:C. A part of thepreparation characterized in panel A was incubated with 5 x104U/ml human thrombin for 4 hr at 37’C prior to gel filtration.(C) Thrombin-inactivated VIII:C. A part of the preparation charac-terized in panel A was incubated with 102U/mI human thrombinfor 4 hr at 37’C prior to gel filtration.
C
60 80 100 120
Elution Volume (ml)
0.01 U/mI Thrombin
C A
I �ti
, “a..0- -0- � -0- -0 - a’ “o- -a- ‘a
THE FACTOR VIII COMPLEX 3
molecular weights of 85,000, 88,000, and 93,000. The
electrophoretic properties did not change when pun-
fled VIlI:C was reduced with 2-mercaptoethanol, but
proteolytic cleavage to smaller proteins could be
accomplished by incubation with thrombin, purified
factor Xa, or activated protein C.2#{176}The purified bovine
VIII:C had no platelet-aggregating activity. Antibod-
ies to the VIII:C protein raised in rabbits inhibited
bovine VIII:C procoagulant activity but had no effect
on the platelet-aggregating activity of bovine plasma.
This observation strongly suggests that the coagulant
protein is separate from the protein responsible for
platelet-aggregating activity.
Human factor VIII procoagulant protein has not
been purified to homogeneity and it may be difficult to
obtain the necessary volume of human plasma that has
been collected in a way that reduces the likelihood of
VIII:C modification in vitro. A major unresolved
problem in VllI:C purification is the poor yield
obtained with all standard methods. For example, only
0.4 mg of bovine VIII:C protein was obtained from
each 125-liter batch of specially collected bovine plas-
ma, an overall VlII:C recovery that was estimated to
be l%.20
We have recently completed studies of VlII:C sepa-
rated from VIIIR and other human plasma proteins by
an immunoadsorbent technique.’7 While this VIII:C is
not stable in the absence of added bovine serum
albumin or similar proteins that prevent loss of VIII:C
from very dilute solutions, the preparation can be
studied by both VIII:C functional assays and
VlII:CAg measurements to determine properties of
VIIt:C when it is separate from VIIIR. Although it is
difficult to exclude the possibility that the protein was
modified during purification, protease inhibitors were
present during the procedure, and the purified VIII:C
had the same ratio of functional to immunologic
activity, as did the plasma from which it was
prepared.
The estimated molecular weight of the human
factor VIII procoagulant protein separated in this
manner is 285,000.21 This value has been calculated
from the properties of VIII:C on Sephadex G-200 gel
filtration (Fig. 2A), from which Stokes’ radius can be
estimated by comparison with standards, and sucrose
density gradient centrifugation (8.25). These proper-
ties of unactivated VIII:C are similar to those recently
obtained for bovine factor V by Nesheim and cowork-
ens.22 The correspondence is not surprising since the
cofactor role of factor V in prothrombin activation is
like that which VIII:C plays in factor X activation.
Both proteins are highly asymmetric molecules that
are activated by thrombin proteolysis.
Factor VIII procoagulant activity is not inactivated
60 80 100 120
Elution Volume (ml)
0.0005 U/mI Thrornbin‘4,
when the intact factor VIII complex is incubated with
reducing agents, e.g., 0.05 M 2-mercaptoethanol.7
This stability is striking when compared to the rapid
loss of nistocetin cofactor activity in these experiments.
Intact thiol groups do appear to have an important role
in VIII:C function, however, and a variety of thiol
inhibitors, including the specific reagent, p-chloromer-
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
1.0
II,-.
>
U4I’-z4_a
DI
0.5
0
NORMAL (22)
S #{149}
C’
C #{149}
. #{149}#{149}
SC #{149}
:.
t t . U I
0 0.5 1.0 1.5
VON WILLEBRAND’S DISEASE (18)
a
aaaa �
ta a
IaI I I
0 0.5 1.0 1.5
HEMOPHILIA (97)
�o;o; �
0 �& � 00 o 0 0
FACTOR �:C ANTIGEN (Units/mI)
Fig. 3. The relationship of factor VIII procoagulant activity
and VIII:CAg in normal individuals and in patients with hemophiliaand von Willebrand’s disease. In the hemophilia panel, the largecircle at the origin represents 43 plasma samples with no detect-able Vlll:C or VIll:CAg. The large circle with an asterisk represents18 plasmas that had <0.01 U/mI VllI:C and 0.01-0.06 U/mIVlll:CAg.
4 LEON W. HOVER
curibenzoic acid, inactivate VIII:C.23 Factor VIII
procoagulant activity is also affected by pH and
calcium concentration. VlIl:C is most stable between
pH 6.9 and 7.2, and a marked loss occurs below pH 6
and above pH 8. The very low VllI:C content of
EDTA and ion-exchange resin-treated plasmas mdi-
cate that plasma cation concentration is also impor-
tant.24
Although neither human nor bovine VIll:C has
been purified in sufficient quantity for carbohydrate
analysis, there is indirect evidence in both cases that
the molecule contains carbohydrate residues. Concon-
avalin-A-agarose binds the purified VIII:C in these
preparations, and VIIl:C procoagulant activity is
eluted by the sugar, a-D glucopyranoside.’7’2#{176}
The limited success of VIll:C purification has made
it necessary to express the plasma content of this
protein by reference to standardized pools of human
plasma collected and stored in a way that reduces the
likelihood of VIlI:C activation or loss. Thus, all
VIII:C and VIII:CAg values are arbitrary measures
that express a ratio. They do not have any molecular
interpretation at the present time. One can, of course,
estimate the amount of protein that corresponds to a
“normal plasma level” of I U/ml. From measure-
ments ofthe intact human factor VIII complex’8’25 and
the proportion of protein that is VIlI:C,6 the value is
approximately 50 ng/U. A similar value can be
obtained from the specific activity of apparently
homogeneous bovine VIII:C (4500 U/mg).2#{176} It must
be recognized, however, that the estimated plasma
concentration, 222 ng/ml, is incorrect if the purified
VIll:C includes protein that has been either activated
or inactivated.
Immunologic Properties
Human anti-VIlI:C, obtained from multitransfused
hemophilic patients that develop inhibitors and from
rare individuals who form autoantibodies that macti-
vate VIIl:C, do not form detectable immunoprecipi-
tates with VIII:C or with the factor VIII complex.
Nevertheless, these sera can be used to detect VIII:C
antigen determinants by antibody neutralization
assays26 and by more sensitive immunoradiometric
assays for VIII:CAg.’5’2729
VIlI:CAg immunoassays require radiolabeled hu-
man anti-VIlI:C that has been purified by preparation
of immune complexes followed by their separation and
dissolution at low pH. The antibodies have been
obtained from high-titer sera (greater than 1000
Bethesda units/ml), and there does not appear to be
any consistent difference between hemophilic and
spontaneous anti-VIII:C. To date, similar results have
been obtained with two different assay methods: fluid
phase and two-site solid-phase separations.’5’2729 The
sensitivity of VlIl:CAg assays is 0.01-0.03 U/mI in
most laboratories, and the coefficient of variation for
these assays is ca. l0%.’�
In general, there is excellent correlation between
plasma factor VIII procoagulant activity and
VIII:CAg content (Fig. 3). VIII:CAg determinants
are more stable than VIII:C activity, however, and this
is striking in the case ofserum in which the VIII:CAg
values are 6O9’o-8O% of those in the corresponding
1.5
I t t U
0 0.5 1.0 1.5
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
THE FACTOR VIII COMPLEX 5
plasma, even though there is no residual VIII:C activi-
ty.2729 Careful studies carried out with purified
VIII:C and human ct-thrombin demonstrated a dose-
dependent loss of VIII:CAg reactivity at thrombin
concentrations below 0.1 NIH U of thrombin/mI-to
60% of the original value-and qualitative changes in
VIlI:CAg determinants upon exposure to higher
thrombin concentrations.2’ These experiments proba-
bly overestimated the effect of thrombin on VIll:C,
since they were carried out in purified systems and the
usual plasma protein inhibitors were not present. It is
clear that thrombin has a detectable effect on
VIII:CAg, but that this measure is much more stable
than procoagulant activity.
Synthesis
Despite many investigations, the site of VIII:C
synthesis is not known. Both transplantation and
perfusion studies strongly suggest that VIII:C is
released by the liver under some conditions.3#{176}33 These
studies, carried out using standard procoagulant
assays, are not definitive, however, and it will be
important to verify these observations by immunoas-
say data and protein synthesis studies. Even though
there is strong suspicion that the liver plays an impor-
tant role in VIlI:C production, the normal or increased
VIII:C values in severe hepatic disease strongly
support the concept that there are extrahepatic
sources of this protein. In any case, it is not yet known
what cell type is responsible for VIII:C synthesis.
Function
It is generally agreed that VIIl:C accelerates blood
coagulation by its cofactor role in the enzymatic
activation of factor X by factor IXa. In the presence of
phospholipid and calcium, VIII:C markedly enhances
this reaction; in the absence of factor IX,, it does not
have any intrinsic capacity to activate factor X.3436
Native VIII:C may not participate in this reaction,
however, and VIlI:C activity is enhanced when plasma
or factor VIII concentrates are incubated with dilute
thrombin.37’38 It is likely, in fact, that thrombin activa-
tion is essential for VIII:C activity.36
It has been presumed that thrombin activates
VIII:C by a proteolytic modification, and this effect
has now been demonstrated.20’2’ Incubation with
thrombin appears to cause a cleavage in each of the
protein chains of bovine VIII:C,2#{176} and thrombin-
activated human VIII:C has gel filtration and ultra-
centrifugation properties of a 1 1 6,000-dalton protein,
in contrast to the 285,000 value calculated for the
unactivated molecule (Fig. 2B).2’ Factor VIII proco-
agulant activity is inactivated by higher concentra-
tions of thrombin-or more prolonged exposure to the
enzyme-and a further shift is noted in the gel filtra-
tion properties of the (nonfunctional) protein
measured by VIII:CAg (Fig. 2C).2’ Thus, thrombin-
mediated activation and inactivation are associated
with proteolytic modifications of VIII:C structure.
FACTOR VIll-RELATED PROTEIN:
VON WILLEBRAND FACTOR
Biochemical Properties
As the bulk of the factor VIII complex consists of
factor VI I I-related protein (V I I I R, von Willebrand
factor), data obtained for bifunctional factor VIII
reflect VIIIR properties.39�’ In all of these studies, the
most striking property of purified factor VIII is its
very large size. Agarose gel filtration separations
suggested a molecular weight greater than I 06,3940 and
the molecular weight obtained by sedimentation equi-
librium studies carried out in 6 M guanidine was
1.12 x 106.41 Although highly purified human factor
VIII is not dissociated by 6 Mguanidine or 1% sodium
dodecyl sulfate (SDS), subunits can be detected when
VIIIR is reduced with 2-mercaptoethanol or dithio-
threitol. A single band is detected on SDS poly-
acrylamide gel electrophoresis, and the subunits have
an estimated molecular weight of 195,000-
240,000.� � 8,39-41
Studies carried out with highly purified factor VIII
may be misleading, however, for they examine only
the small fraction of the molecules (ca. 5%-10%) that
are not lost during the series of separations. It is now
apparent that factor VIII-related protein is, in fact, a
heterogeneous population of multimers that have a
range of molecular weights from ca. 850,000 to over
12 x 106. This property first became apparent when
the technique of crossed immunoelectrophoresis iden-
tified VIIIR heterogeneity.42 Zone electrophoresis in
agarose did not resolve the different forms, however,
and the population of multimers was not recognized
until VIIIR was examined in agarose or agarose/
acrylamide gels in the presence of sodium dodecyl
sulfate.7’43 These studies of porcine and human VIIIR
were carried out with purified materials, and it was
possible that the apparent multimeric pattern could
have been caused by aggregation that occurred during
purification. Subsequent analysis of unmodified
normal human plasma has demonstrated that VIIIR
circulates as a population of very large multimers and
that the size distribution is not an artifact induced by
purification methods, freezing, or calcium chela-
tion”45 (Fig. 4).
The smallest polymers detected in normal human
plasma have an apparent Mr of 0.85 x I �6#{149}This would
appear to be a disulfide-bonded tetramer of the basic
ca. 200,000 subunit. The larger forms have M� indicat-
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
6 LEON W. HOVER
2.8 x i06-1.9 x 106-
0.95 x iO6�__
1gM NJ HEM VWD VWDI It
Fig. 4. The polymer pattern of human plasma VIIR analyzed by SDS-agarose electrophoresis. The migration of 1gM and 1gM polymers
is indicated on the left and their molecular weights are noted. The Vlll:R in plasma samples was identified by autoradiography afterincubation with tThl�Iabeled rabbit anti-VIIIR:Ag according to the method of Hoyer and Shainoff.” From the left. these plasmas are from anormal individual and patients with severe von Willebrand’s disease (VIIIR:Ag < 0.01 U/mI). severe hemophilia. type I von Willebrand’sdisease. and type II von Willebrand’s disease.
ing that they are composed of an integral number of
these tetramers, i.e., they have M� of 1 .7 x 1 06, 2.5 x
106, 3.4 x 106, etc. (Fig. 4). Serum obtained from
blood clotted in glass tubes at 37#{176}Chas the same
population of VIIIR.44 As many as 8 separate bands
can be detected in most normal plasmas, and there is
in addition a population of poorly resolved VIIIR with
Mr of ca. 8-1 2 x 106.
Purified human VIIIR has also been analyzed by
standard biochemical methods. It is a glycoprotein
containing 5%-6% carbohydrate, and hexose, hexos-
amine, and sialic acid have been specifically identi-
fled.39’4’ The amino acid composition has also been
determined in these studies: methionine, tyrosine, and
tryptophane values are relatively low and there are no
free sulfhydryl groups.39’
The amount of VIIIR protein in plasma has been
calculated from the specific activity of highly purified
factor VIII’8’39’ and from immunoradiometric assays
of plasma in which purified VIIIR was used as a
standard.25 The estimated values for VIIIR protein in
normal human plasma have been between 5 and 10
sg/ml. This is approximately 100 times greater than
the concentration of VIII:C protein.
Immunologic Properties
Rabbits immunized with purified human factor
VIII form useful immunoprecipitating antibodies.�
Although the sera vary in their ability to inactivate
VIII:C, they all form immunoprecipitates with VIIIR
and they inactivate plasma ristocetin cofactor activity
as well as other measures of VIIIR-platelet interac-
tion.2’47’48 They are monospecific in immunoprecipitin
assays if prepared with sufficiently purified factor
VIII, but absorption with VIIIR-depleted plasma
fractions is often necessary. A number of different
assays have been used to detect and quantify this
factor Vill-related antigen. Laurell electroimmunoas-
say, counter immunoelectrophoresis, and radioimmu-
noassay all give similar results,”4’46’49 and antisera
prepared in different laboratories have had generally
consistent properties in these immunochemical assays.
These rabbit antisera vary in their effect on human
VIII:C activity.50 The properties of the material used
for immunization appear to be very important in this
regard, and the VIII:CAg content depends on the
purification method. Although small amounts of anti-
VIII:CAg may be present in some sera (in addition to
the anti-VIIIR:Ag) they do not affect the immuno-chemical measurements. The interaction of anti-
VIIIR:Ag with the intact factor VIII complex will also
inactivate VIII:C, and anti-VIIIR:Ag coupled to
agarose removes both VIIIR and VIII:C from plas-
ma.’6’7
In general, there is a good correlation between the
factor VIII procoagulant activity and the factor VIII-
related antigen concentration in normal human plas-
mas (Fig. 5). Parallel increases in VIII:C and
VIIIR:Ag have been noted in plasmas from patients
with a wide range of nonhematologic diseases and
from normal individuals subjected to physiologic stim-
uli.”�
von Willebrand Factor Activity
Factor-VIII-related protein has a central role in
normal platelet function, a property that has been
designated “von Willebrand factor” activity because it
is deficient in patients with von Willebrand’s disease.’
The prolonged bleeding time in these patients is
presumed to be due to the reduced plasma VIIIR
content and it is corrected by transfusion of VIIIR-
rich cryoprecipitate. Two in vitro platelet assays are
usually abnormal in von Willebrand’s disease: ristoce-
tin-induced platelet agglutination and retention of
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
S NORMAL (59)
� VON WILLEBRAND’S DISEASE (23)0 HEMOPHILIA (42)
S
(5.
>-
>
0_ D 0.5
0U
0�
I#{149}5�
S S
S
0
000 0 � 0
0 � o� 0 o0g00 o gg � Q:� 0 �b o 0 00
6 o’s
0
1.5
0
i.O 2.5
FACTOR yin-RELATED ANTIGEN (Units/mI)
Fig. 5. The relationship of factor VIII procoagulant activity and factor VIll-related antigen in the plasmas of normal individuals andpatients with hemophilia or von Willebrand’s disease.
THE FACTOR VIII COMPLEX
S#{149}.
S
S. �
.: �#{149}#{149}#{149}:S #{149}.
%S #{149}.#{149}#{149}#{149}SS S #{149}
S #{149}
platelets in glass bead columns.’ Both are corrected
when purified normal factor VIII is added to blood of
a patient with severe von Willebrand’s disease.
Although the way in which a prolonged bleeding time
is a consequence of the reduced level of factor VIII
protein is not yet known, there is a good correlation
between this abnormality and reduced levels of plasma
VIIIR:Ag and ristocetin cofactor measurements.5’
The use of ristocetin for in vitro assessment of
VIIIR function has become widely adopted in a rather
short period of time. The initial observation-reduced
or absent platelet aggregation when ristocetin was
added to platelet-rich plasma (PRP) from patients
with von Willebrand’s disease52-provided a simple
measure that many laboratories have incorporated
into the routine evaluation of patients with bleeding
disorders. Unfortunately, the assessment of ristocetin-
induced aggregation in patient PRP has limitations as
a diagnostic technique. In addition to its qualitative
nature, normal or reduced aggregation at one or more
ristocetin concentrations, it may be falsely positive in
some patients with primary platelet disorders and it is
not sufficiently sensitive to detect mild or moderate
von Willebrand’s disease.53 While abnormal aggrega-
tion in von Willebrand’s disease is the consequence of
a plasma deficiency and can be corrected by addition
of normal VIIIR, the phenomenon also requires the
presence of a normal platelet surface protein that is
deficient in Bernard-Soulier syndrome.54
Ristocetin-induced platelet aggregation is more
critically examined by assays in which dilutions of
plasma are tested with washed normal platelets and a
fixed concentration of ristocetin. These ristocetin
cofactor assays can be done with freshly washed plate-
lets or with formaldehyde-fixed platelets that remain
satisfactory as test reagents for several weeks if kept in
the refrigerator. The rate of ristocetin-induced aggre-
gation is related to the amount of plasma that is
added, and the value can be obtained from the aggre-
gometer tracing or by measurement of the time
required by detectable platelet agglutination.55 The
different methods give similar results under most
conditions.
Although there is a good correlation of ristocetin
cofactor activity in vitro and presumed von Willebrand
factor activity in vivo, as judged by freedom from
abnormal bleeding and by bleeding time measure-
ments,5’ there are exceptions. For example, the
VIIIR:RC value may become normal in von Wille-
brand’s disease during pregnancy and in inflammatory
states, or after transfusion with factor VIII, even
though the bleeding time remains prolonged.56’57 In
addition, patients with a variant form of von Wille-
brand’s disease have long bleeding times in spite of
low-normal ristocetin cofactor values and increased
reactivity when ristocetin is added to their platelet-
rich plasma.45’58 Thus, the value of ristocetin cofactor
assays as in vitro measures of VIIIR function must not
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
8 LEON W. HOVER
obscure the fact that they do not always reflect in vivo
biologic function. In this regard, it should also be
emphasized that VIIIR:Ag and ristocetin cofactor
assays measure different properties of the VIIIR
protein. While immunoassays detect all VIIIR mole-
cules with specific antigenic determinants, the protein
does not always have biologic activity. Moreover, arti-
factually increased immunoassay values are obtained
for the smaller VIIIR polymers when they are
compared to whole plasma standards by the Laurell
electroimmunoassay method. In contrast, ristocetin
cofactor measurements only identify VIIIR protein
that can interact with platelets, and this capacity is
limited to the larger polymers.59 Thus, plasmas or
purified proteins that lack the larger VIIIR polymers
will have a very low ratio of VIIIR:RC to VIIIR:Ag,
and, conversely, material that is relatively enriched in
large forms will have higher values for ristocetin
cofactor activity than VIIIR:Ag, when compared to
the whole plasma standard.
Radiolabeled purified VIIIR binds specifically to
platelet membranes, and it has been suggested that
discrete receptor sites are present.6#{176}62 Ristocetin,
which also binds to the platelet membrane,63 appar-
ently makes this platelet receptor available to VIIIR
and enhances its binding. In fact, it has been shown
that increasing concentrations of ristocetin cause more
VIIIR molecules to bind to the platelets.60’6’ A correla-
tion between VIIIR binding to platelets and ristocetin-
induced platelet aggregation has also been demon-
strated.62 VII1R polymer heterogeneity complicates
these analyses, however, for it is now recognized that
only large VIIIR bind to platelets in the presence of
ristoceti n �
Several kinds of data suggest that VIIIR carbohy-
drate is important in the factor-VIII-platelet interac-
tion and in ristocetin-induced aggregation. Initial
studies indicated that sialic acid removal reduced
ristocetin-induced platelet aggregation by 65%;64 other
studies found no change in reactivity even though the
sialic acid was removed.65 The latter studies suggested
that oxidation of the penultimate galactose modified
ristocetin-induced platelet aggregation, and there was
a 90% reduction of VIIIR:RC function when these
residues were oxidized. Subsequent reversal of the
reaction by galactose reduction caused full regenera-
tion of VllIR:RC function. No change in VIII:C
activity was noted when intact factor VIII complexes
were modified in a way that removed or oxidized these
carbohydrate groups.64’65
Carbohydrate residues are also important in VIIIR
intravascular survival. The asialo-derivative is cleared
by the rabbit liver with a T’/2 of 5 mm; the value for
normal VIIIR is 240 mm.64 Other studies have demon-
strated that the rabbit liver lectin that specifically
binds asialoglycoproteins binds to the asialo-deriva-
tive, not the native or asialo-agalacto-VlIlR.66
Synthesis
Immunofluorescent studies have identified
VIIIR:Ag in endothelial cells of arteries, arterioles,
capillaries, and veins throughout the body, as well as
in megakaryocytes and platelets.67’68 In addition, both
VIIIR:Ag and VIIIR:RC have been identified in the
medium from cultured human umbilical cord endothe-
hal cells.69’70 Direct evidence of VIIIR synthesis by
endothelial cells has also been obtained in tissue
culture.69 Factor VIII coagulant activity was not
detected in these culture media, however. While it is
possible that VIII:C might have been inactivated by a
protease present in the media, VIII:CAg was also
undetectable.7’ Thus, either the culture conditions
and/or cell source (umbilical cord veins) are unsatis-
factory for VIlI:C synthesis or VIIl:C is synthesized
by a different cell type.
THE INTERACTION OF VIII:C and VIIIR IN THE
FACTOR VIII COMPLEX
Although VIll:C and VIIIR have very distinct
properties, it would be an oversimplification to suggest
that they have no relationship and that they are simply
two proteins that happen to copurify. Several observa-
tions indicate that in plasma they interact through
noncovalent bonds to form a complex. For example,
the concentrations of the two proteins are closely
correlated in normal plasma and in most (nonhemato-
logic) disease states (Fig. 5)#{149}l.14.46.72 In contrast, there
is no correlation of either protein with factor V activi-
ty.72 The interaction between the components remains
intact when VIIIR interacts with heterologous anti-
bodies (e.g., rabbit anti-(whole) factor VIII) and
VIII:C is included in the immune complex. Immuno-
precipitates obtained with rabbit anti-VIIIR have
coagulant activity73 and elicit anti-VIll:C when
injected into other rabbits.74 Moreover, anti-VIIIR
coupled to agarose removes both VIII:C and VIIIR
from plasma.16”7 IfVIIl:C and VIIIR did not interact,
one would not expect the VIIJ:C to remain with the
immunoadsorbent. An alternative reason for a loss of
VIIl:C from the plasma, direct binding to unrecog-
nized anti-VIII:C in the rabbit antiserum, can be
discarded since the VIII:C can be recovered from the
beads by a modest increase in the ionic strength that
does not affect immunologic reactions.’7
Human antibodies to factor VIII have a different
effect. Although agarose-bound antibodies remove
VIII:C procoagulant activity (and VIIl:CAg) from
plasma, plasma VIIIR:Ag and ristocetin cofactor
activity are unaffected.’6 The difference between the
human and the heterologous antibodies in these exper-
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
THE FACTOR VIII COMPLEX 9
iments is likely to be due to the very different concen-
trations of VIII:C and VIIIR protein in plasma. Since
less than 1% of the protein in the factor VIII complex
is VIII:C, the amount of VIIIR removed with VIII:C
may be so small that the assays cannot detect a change
in VIIIR concentration. In addition, the factor VIII
complex appears to be “destabilized” when VIII:C
interacts with human antibodies. The basis for this
suggestion is found in studies of soluble complexes
obtained by incubating human anti-VIII:C with
normal plasma. Immune complexes were detected in
late-eluting fractions, as well as at the void volume,
even though VIII:C was limited to the void volume
fractions when the antibody was not present.75
A relatively high-affinity interaction between
VIII:C and VIIIR can also be inferred from the effect
of reducing agents on plasma VIII:C. After plasma is
exposed to low concentrations of dithiothreitol or 2-
mercaptoethanol, VIII:C has the properties of a rela-
tively small protein on sucrose density centrifugation,
agarose gel filtration, and ethanol precipitation.76
While this change could be due to a direct effect of the
reducing agents on VIII:C properties, they return to
“normal” if hemophilic plasma (VIIIR free of VIII:C
activity) is added. This sequence strongly suggests
that the presence of intact VIIIR modifies the proper-
ties of the VIIl:C protein in standard separation tech-
niques.
The data summarized above suggest that VIII:C
and VIIIR interact in some way, but we have very
little information about the way in which this complex
is formed. Nevertheless, the susceptibility to dissocia-
tion by high salt buffers is indirect evidence that
electrostatic forces are important. The biologic impor-
tance of the plasma interaction is also uncertain, but
VIII:C instability in the absence of VIIIR (or in
plasmas that have relatively reduced VIIIR) suggests
that complex formation protects VlII:C from proteo-
lytic inactivation.77
THE FACTOR VIII DEFICIENCY DISEASES
Hemophilia A
Although the low factor VIII concentration in
plasma has prevented biochemical studies in classic
hemophilia (hemophilia A), immunologic techniques
have begun to define the molecular defect. These
studies have attempted to distinguish diminished
production of normal factor VIII from synthesis of
nonfunctional protein.
Two kinds of immunologic studies must be clearly
differentiated as this question is considered. The initial
work, done with human antibodies by the technique of
antibody neutralization, identified “nonfunctional but
antigenically cross-reacting AHF-like protein” in 10%
of hemophilic plasmas.26’78’79 The plasmas were desig-
nated cross-reacting material positive (CRM + ), and
they are now known to have normal levels of V I I I :CAg
even though VIII:C is very low (2%-lO% of normal).’5
Shortly thereafter, quantitative immunoassays were
carried out with rabbit antisera to human factor VIII.
These studies identified (by quantitative immunopre-
cipitin measurements) normal levels of “factor-VIlI-
like protein” in all hemophilic plasmas.46 Moreover, all
hemophilic plasmas neutralized the VIII:C inactivat-
ing properties of this rabbit antiserum. Plasmas from
patients with severe von Willebrand’s disease did not
form immunoprecipitates nor did they neutralize the
anti-VIII:C activity.46 These studies, and others
carried out with heterologous antisera, led to the
concept that nonfunctional but immunologically cross-
reactive material is present in all hemophilic plas-
mas.’4’46 It is now recognized that immunochemical
assays using heterologous antisera-whether done by
immunoprecipitation, hemagglutination, or radioim-
munoassay-detect VIIIR:Ag, not antigens related to
VIII:C function. Thus, they do not demonstrate an
intact factor VIII complex in hemophilia A, only
normal VIIIR:Ag synthesis as one might expect from
the normal bleeding time and the normal values of in
vitro assays of von Willebrand factor activity. It was
subsequently shown that VIIIR purified from hemo-
philic plasmas cannot be distinguished from normal
VIIIR by standard biochemical methods.�’#{176} Thus,
VIIIR:Ag measurements can be used as an assay that
distinguishes hemophilia from most forms of von
Willebrand’s disease, but they do not provide any
information about the nature of hemophilia. They
measure the product of a different (autosomal) gene.
Immunologic study of hemophilia has become possi-
ble, however, as human anti-VIII:C have been used in
quantitative immunoradiometric assays for VIII:C
antigenic determinants.’5’27 This technique has permit-
ted both qualitative and quantitative evaluation of
VIIIC:Ag and a number of studies have been
published during the past 2 yr.’5’27�29’8#{176}They have
identified several different patterns of VlII:C and
VIII:CAg in hemophilic plasmas (Fig. 3).
There is no detectable VIII:CAg in most plasmas
from patients with severe hemophilia (VIll:C <1% of
normal). In one-fourth of these plasmas (33 of I 34)
there are low VIII:CAg levels-usually I%-l0% of
normal, but rarely as much as 28%-even though
there is no detectable VIII:C coagulant activity.’5’27
29,80,81 Variable VIII:CAg levels are present in plasmas
of patients with mild or moderate hemophilia; usually
there is slightly more immunoreactive material than
VIII:C. A more extreme difference is observed for the
small group of hemophilic plasmas that have normal
VIII:CAg even though the coagulant activity is low.
These plasmas are from the same patients in which
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
10 LEON W. HOVER
CRM+ hemophilia can be identified by antibody
neutralization assays.26
Thus, patients with hemophilia have VIII:C defi-
ciency transmitted by X-chromosome inheritance and
they have normal VIIIR synthesis and function.
Nonfunctional VIII:C-like molecules are synthesized
by some hemophilic patients, and plasma concentra-
tions of immunoreactive protein are normal in rare
instances. These patients have an X-chromosome
mutation that modifies VIII:C structure. The absence
ofdetectable VIII:C protein in the remaining patients
may reflect a structural defect that is so severe that
antigenic reactivity is lost (as well as coagulant func-
tion) or it may indicate that there is no protein in the
plasma.
The evolving understanding of the molecular defect
in hemophilia has been based on new methods
(VIII:CAg and VIIIR:Ag measurements) that have
supplemented standard coagulation assays. This has
also improved our ability to provide informed genetic
counseling for families affected by this disease. It is
now widely recognized that hemophilia carrier detec-
tion has been facilitated by the combined measure-
ments of VIII:C and VIIIR:Ag, and most (>85%)
hemophilia carriers can be identified when the two
assays are done in laboratories that have excellent
assay quality control and sufficient experience with
reference populations of normal and genetically obli-
gate carriers.82’83 Carrier women have normal
VIIIR:Ag levels; VIII:C is reduced since only half of
their X-chromosomes (on the average) direct normal
VIII:C synthesis. The maintenance of an abnormal
ratio suggests that the physiologic influences that
affect the factor VIII complex act by modifying the
production, release, and metabolism of the two
proteins in a consistent way.
VlII:CAg stability in the presence ofamniotic fluid
had led to a further advance in genetic counseling, and
prenatal diagnosis of hemophilia is feasible by immu-
noassay analysis of fetal blood samples obtained by
fetoscopy at I 8-20 wk of gestation.84’85 In collabora-
tive studies carried out with Dr. M. J. Mahoney of the
Department of Human Genetics of Yale University
School of Medicine, immunoradiometric assay for
VIII:CAg (and VIIIR:Ag as a control protein) has
excluded-or accurately identified-hemophilia in
utero for 35 fetuses tested through October 1 , I 980.86
von Willebrand�s Disease
Immunologic and biochemical assays have also clar-
ified our understanding ofvon Willebrand’s disease. In
this case, the hemostatic disorder is transmitted by an
autosomal locus that affects VIIIR structure and
concentration. Qualitative and quantitative VIIIR
defects have been identified in this disease; the
reduced VIII:C levels appear to be secondary.
In its most frequent form, von Willebrand’s disease
is a mild or moderate bleeding disorder in which all of
the components of the factor VIII complex are
reduced in quantity and the bleeding time is
prolonged.”87 Plasma VIII:C and VIII:CAg may be
slightly higher than VIIIR:Ag and VIIIR:RC, but the
values are usually similar (Figs. 3 and 5). The VIIIR
multimer pattern appears to be normal since all of the
VIIIR polyers are reduced in quantity.45 The bleeding
time prolongation is variable, but it is usually asso-
ciated with reduced ristocetin cofactor activity.5’
While these patients all have genetic defects that
affect VIIIR, the hemostatic defect may be quite
inconsistent and the results of plasma assays and
bleeding time measurements may vary considerably
from time to time.88
Severe von Willebrand’s disease is a distinct (and
unusual) condition in which individuals have very low
levels of both factor VIII components, a markedly
prolonged bleeding time, and a major bleeding diathe-
sis. Family studies often demonstrate that these
patients are homozygous offspring of parents with
mild, or asymptomatic, von Willebrand’s disease.
Although VIIIR:Ag and VIIIR:RC levels are usually
less than 1% of normal in severe von Willebrand’s
disease (Fig. 5), some VIIIR:Ag can usually be
detected by sensitive assay methods.89
It is not certain why VIIl:C is low in von Wille-
brand’s disease, for these patients have the genetic
capacity to synthesize this protein. Recent studies
suggest that normal VIIIR protects VIII:C from mac-
tivation, and it is possible that VIIIR deficiency
permits accelerated VIII:C inactivation in vivo.77 It is
also possible that plasma VIIIR levels may, in some
poorly understood way, have an effect on VIII:C
synthesis or its release into the plasma.33 It is likely
that an understanding of the low baseline VIII:C level
will clarify the mechanism of the delayed rise and
prolonged survival of VIII:C (and VIII:CAg) after
transfusion in von Willebrand’s disease.’ At the pres-
ent time, one can simply suggest that normal (trans-
fused) VIIIR may stabilize VIlI:C or it may directly
influence VIII:C synthesis or release.
Most patients with von Willebrand’s disease have
similar levels of the different factor VIII properties-
this has been designated the “classical” pattern. Other
patients have been found to have nonfunctional VIIIR,
and these individuals have normal or slightly reduced
VIII:C and VIIIR:Ag, very low VIIIR:RC, and a
prolonged bleeding time. The VIIIR:Ag pattern is
abnormal on crossed immunoelectrophoresis, and this
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
THE FACTOR VIII COMPLEX 11
has suggested that there is an abnormal and nonfunc-
tional VIIIR.90’9’ It is now apparent that the more
rapid VIIIR migration reflects an increased propor-
tion of small VIIIR and an absence of the largest
multimers (Fig. 4)#{149}45.92 The shift in multimer distribu-
tion changes the VIIIR electrophoretic mobility, since
agarose electrophoresis separates large molecules
according to size as well as charge. The important
consequence of the abnormal polymer distribution is
impaired hemostasis in the variant form of von Wille-
brand’s disease. Platelet binding, ristocetin cofactor
activity, and bleeding time correction all require the
large VIIIR forms and their deficiency causes a bleed-
ing diathesis.45’59’92 This variant form of von Wille-
brand’s disease has been designated “type II” by some
investigators to distinguish it from the classical “type
I” pattern in which there is a normal polymer pattern
and a similar reduction in all components of the factor
VIII complex.
A distinction has recently been made between two
subtypes of type II von Willebrand’s disease and the
abnormal VIIIR polymer patterns are slightly
different. In addition, PRP from patients with the
“type IIB” disorder has increased reactivity when
tested for ristocetin-induced platelet aggregation and
plasma VIIIR:RC is normal or minimally reduced.45’58
In contrast, patients with the more common “type
lIA” disease have markedly reduced ristocetin cofac-
tor activity and there is no aggregation when ristocetin
is added to their PRP.
The importance of carbohydrate groups in VIIIR
function has been considered in a previous section. It
is, therefore, not surprising that reduced VIIIR carbo-
hydrate has been identified in some patients with von
Willebrand’s disease. VIIIR subunits prepared from
these plasmas do not have detectable carbohydrate
when tested after SDS-polyacrylamide gel electropho-
resis, and the VIIIR sialic acid content is reduced.93’94
While a carbohydrate abnormality may be responsible
for some instances of nonfunctional VIIIR, another
series suggests that it is an uncommon form of von
Willebrand’s disease.95
REFERENCES
I. Hoyer LW: von Willebrand’s disease. in Spaet TH (ed):
Progress in Hemostasis and Thrombosis. vol 3. New York, Grune &
Stratton, (976, pp 23 1-287
2. Bouma BN, Wiegerinck Y, Sixma ii, et al: Immunological
characterization of purified anti-haemophilic factor A (factor VIII)
which corrects abnormal platelet retention in von Willebrand’s
disease. Nature (New Biol) 236:104-106, 1972
3. Forbes CD, Prentice CRM: Aggregation of human platelets
by purified porcine and bovine antihaemophilic factor. Nature
(New Biol) 241:149-150, 1973
4. McKee PA, Andersen JC, Switzer ME: Molecular structural
studiesofhuman factor VIII. Ann NY Acad Sci 240:8-33, 1975
5. Legaz ME, Weinstein Mi, Heldebrant CM, Davie EW: Isola-
tion, subunit structure and proteolytic modification of bovine factor
VIII. Ann NY Acad Sci 240:43-61, 1975
6. Switzer ME, McKee PA: Studies on human antihemophilic
factor. J Clin Invest 57:925-937, 1976
7. Counts RB, Paskell SL, Elgee SK: Disulfide bonds and the
quaternary structure of factor Vill/von Willebrand factor. J Clin
Invest 62:702-708, 1978
8. Weiss Hi, Kochwa 5: Molecular forms of antihaemophilic
globulin in plasma, cryopreciptate and after thrombin activation. Br
i Haematol 18:89-100, 1970
9. Owen WG, Wagner RH: Antihemophilic factor: Separation of
an active fragment following dissociation by salts or detergents.
Thromb Diath Haemorrh 27:502-515, 1972
10. Rick ME, Hoyer LW: Immunologic studies of antihemo-
philic factor (AHF, factor VIII). V. Immunologic properties of
AHF subunits produced by salt dissociation. Blood 42:737-747,
I 973
I I . Weiss Hi, Hoyer LW: von Willebrand factor: Dissociation
from antihemophilic factor procoagulant activity. Science
182:1149-1151, 1973
I 2. Poon MC, Ratnoff OD: Evidence that functional subunits of
antihemophilic factor (factor VIII) are linked by noncovalent bonds.
Blood48:87-94, 1976
13. Sussman II, Weiss Hi: Dissociation of factor VIII in the
presence of proteolytic inhibitors. Thromb Haemostas 40:316-325,
I 978
14. Hoyer LW: Immunologic studies of antihemophilic factor
(AHF, factor VIII). IV. RadioimmunoassayofAHFantigen. i Lab
Clin Med 80:822-833, 1972
I 5. Lazarchick i, Hoyer LW: Immunoradiometric measurement
of the factor VIII procoagulant antigen. i Clin Invest 62:1048-
1052, 1978
16. Zimmerman IS, Edgington TS: Factor VIII coagulant activ-
ity and factor VIlI-like antigen: Independent molecular entities. I
Exp Med 138:1015-1020, 1973
I 7. Tuddenham EGD, Trabold NC, Collins IA, Hoyer LW: The
properties of factor VIII coagulant activity prepared by immunoad-
sorbent chromatography. J Lab Clin Med 93:40-53, 1979
18. Olson ID, Brockway WI, Fass DN, Bowie EJW. Mann KG:
Purification of porcine and human ristocetin-Willebrand factor. I
LabClin Med89:1278-1294, 1977
19. Bloom AL: Physiology of factor VIII, in Poller L (ed):
Recent Advances in Blood Coagulation. Edinburgh, Churchill-
Livingston, 1977, pp 141-181
20. Vehar GA, Davie EW: Preparation and properties of bovine
factor VIII (antihemophilic factor). Biochemistry 19:401-410,
I980
21. Hoyer LW, Trabold NC: The effect or thrombin on human
factor VIII. Cleavage ofthe factor VIII procoagulant protein during
activation. I Lab Clin Med 97:50-64, 1981
22. Nesheim ME, Myrmel KH, Hibbard L, Mann KG: Isolation
and characterization of single chain bovine factor V. I Biol Chem
254:508-517, 1979
23. Austen DEG: Thiol groups in the blood clotting action of
factor VIII. Br I Haematol 19:447-484, 1970
24. Weiss HI: A study of the cation- and pH-dependent stability
of factors V and VIII in plasma. Thromb Diath Haemorrh 14:32-
51, 1965
25. Counts RB: Solid phase immunoradiometric assay of factor
VIII protein. Bri Haematol 31:429-436, 1975
26. Hoyer LW, Breckenridge RT: Immunologic studies of anti-
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
12 LEON W. HOVER
hemophilic factor (AHF, factor VIII): Cross-reacting material in a
genetic variant of hemophilia A. Blood 32:962-97 1 , 1968
27. Peake IR, Bloom AL, Giddings JC, Ludlam CA: An immu-
noradiometric assay for procoagulant factor VIII antigen: Results in
haemophilia, von Willebrand’s disease and fetal plasma and serum.
Bri Haematol42:269-281, 1979
28. Holmberg L, Burge L, Ljung R, Nilsson IM: Measurement
of antihemophilic factor A antigen (VIII:CAg) with a solid phase
immunoradiometric method based on homologous non-haemophilic
antibodies. Scand I Haematol 23:17-24, 1979
29. Muller HP, van Tilburg NH, Bertina RM, Terweil JP,
Veltkamp II: Immunoradiometric assay ofprocoagulant factor VIII
antigen (VlII:CAg). Clin Chim Acta 107:1 1-19, 1980
30. Webster WP, Zukoski CF. Hutchin P. Reddick RL, Mandel
ST. Penick GD: Plasma factor VIII synthesis and control as
revealed by canine organ transplantation. Am J Physiol 220:1 147-
1154, 1971
31. Owen CA Jr. Bowie EJW, Fass DN: Generation of factor
VIII coagulant activity by isolated, perfused neonatal pig livers and
adult rat livers. BrI l-Iaematol 43:307-315, 1979
32. Shaw E, Giddings IC, Peake lR, Bloom AL: Synthesis of
procoagulant factor VIII, factor VIII related antigen and other
coagulation factors by the isolated perfused rat liver. Br I Haematol
41:585-596, 1979
33. Bloom AL: The biosynthesis of factor VIII. Clin Haematol
8:53-77, 1979
34. Osterud B, Rapaport SI: Synthesis of intrinsic factor X
activator. Inhibition of the function of formed activator by antibod-
ies to factor VIII and to factor IX. Biochem I 9:1854-1861, 1970
35. Suomela I-I, Blomb#{228}ck M, Blomb#{228}ck B: The activation of
factor X evaluated by using synthetic substrates. Thromb Res
10:267-281, 1977
36. Hultin MB, Nemerson Y: Activation of factor X by factors
1X0 and VIII; a specific assay for factor 1X0 in the presence of
thrombin-activated factor VIII. Blood 52:928-940, 1978
37. Rapaport SI, Schiffman 5, Patch Mi, Ames SB: The impor-
tance of activation of antihemophilic globulin and proaccelerin by
traces of thrombin in the generation of intrinsic prothrombinase
activity. Blood 21:221-235, 1963
38. #{216}sterud B, Rapaport SI, Schiffman 5, Chong MNY: Forma-
tion of intrinsic factor-X-activator activity, with special reference to
the role of thrombin. Br I Haematol 21 :643, 1971
39. Marchesi SL, Shulman NR, Gralnick HR: Studies on the
purification and characterization of human factor VIII. J Clin
Invest 51:2151-2161, 1972
40. Shapiro GA, Andersen IC, Pizzo SV, McKee PA: The
subunit structure of normal and hemophilic factor VIII. I Clin
Invest 52:2198-2210, 1973
41. Legaz ME, Schmer G, Counts RB, Davie EW: Isolation and
characterization of human factor VIII (antihemophilic factor). J
Biol Chem 248:3946-3955, 1973
42. Zimmerman TS, Roberts J, Edgington TS: Factor-VIlI-
related antigen: Multiple molecular forms in human plasma. Proc
NatI Acad Sci USA 72:5121-5125, 1975
43. Fass DN, Knutson GI, Bowie EJW: Porcine Willebrand
factor: A population of multimers. I Lab Clin Med 91:307-320,
I978
44. Hoyer LW, Shainoff IR: Factor-VIlI-related protein
circulates in normal human plasma as high molecular weight
multimers. Blood 55:1056-1059, 1980
45. Ruggeri ZM, Zimmerman TS: Variant von Willebrand
disease. Characterization of two subtypes by analysis of multimeric
composition of factor VIll/von Willebrand factor in plasma and
platelets.I Clin Invest65:1318-1325, 1980
46. Zimmerman TS, Ratnoff OD, Powell AE: Immunologic
differentiation of classic hemophilia (factor VIII deficiency) and
von Willebrand’s disease, with observations on combined deficien-
cies of antihemophilic factor and proaccelerin (factor V) and an
acquired circulating anticoagulant against antihemophilic factor. J
Clin Invest 50:244-254, 1971
47. Meyer D, Jenkins C, Dreyfus M, Larriev Mi: An experimen-
tal model for von Willebrand’s disease. Nature 243:293-294, 1973
48. Baumgartner HR, Tschopp TB, Meyer D: Shear rate depen-
dent inhibition of platelet adhesion and aggregation on collagenous
surfaces by antibodies of human factor VIII/von Willebrand. Br J
Haematol 44:127-139, 1980
49. Zimmerman TS, Hoyer LW, Dickson L, Edgington TS:
Determination of the von Willebrand’s disease antigen (factor
VIlI-related antigen) in plasma by quantitative immunoelectro-
phoresis. I LabClin Med 86:152-159, 1975
50. Kernoff PBA, Rizza CR: The specificity of antibodies to
factor VIII produced in the rabbit after immunization with human
cryoprecipitate. Thromb Diath Haemorrh 29:652-660, 1973
51. Weiss HI, Hoyer LW, Rickles FR, Varma A, Rogers J:
Quantitative assay of a plasma factor, deficient in von Willebrand’s
disease, that is necessary for platelet aggregation. Relationship to
factor VIII procoagulant activity and antigen content. J Clin Invest
52:2708-2716, 1973
52. Howard MA, Firkin BG: Ristocetin-A new tool in the
investigation of platelet aggregation. Thromb Diath Haemorrh
26:363-369, 1971
53. Howard MA, Sawers RI, Firkin BG: Ristocetin: A means of
differentiating von Willebrand’s disease into two groups. Blood
41:687-690, 1973
54. Howard MA, Hutton RA, Hardisty RM: Hereditary giant
platelet syndrome: A disorder of a new aspect of platelet function.
Br Med I 2:586-588, 1973
55. Sarji KE, Stratton RD, Wagner RI-I, Brinkhous KM: Nature
of von Willebrand factor: A new assay and a specific inhibitor. Proc
NatI Acad Sci USA 71:2937-2941, 1974
56. Ratnoff OD, Saito H: Bleeding in von Willebrand’s disease.
N EngI J Med 290:1089, 1974
57. Weiss Hi: Relation of von Willebrand factor to bleeding
time. N Engl I Med 29 1 :420, 1974
58. Ruggeri ZM, Pareti Fl, Mannucci PM, Ciavarella N,
Zimmerman TS: Heightened interaction between platelets and
factor VIlI/von Willebrand factor in a new subtype of von Wille-
brand’sdisease. N Engl I Med 302:1047-1051, 1980
59. Doucet-deBruine MHM, Sixma ii, Over I, Beeser-Visser
NH: Heterogeneity of human factor VIII. II. Characterization of
forms of factor VIII binding to platelets in the presence of ristocetin.
I Lab Clin Med 92:96-107, 1978
60. Kao K-i, Pizzo SV, McKee P: Demonstration and character-
ization of specific binding sites for factor VIlI/von Willebrand
factor on human platelets. I Clin Invest 63:656-664, 1979
61. Morisato DK, Gralnick HR: Selective binding of the factor
VIlI/von Willebrand factor protein to human platelets. Blood
55:9-15, 1980
62. Kao K-i, Pizzo SV, McKee P: Platelet receptors for human
factor VIlI/von Willebrand factor protein: Functional correlation of
receptor occupancy and ristocetin-induced platelet aggregation.
Proc Natl Acad Sci USA 76:5317-5320, 1979
63. Coller BS: The effects of ristocetin and von Willebrand factor
on platelet electrophoretic mobility. J Clin Invest 61:1 168-1 175,
I977
64. Sodetz JM, Pizzo SV, McKee P: Relationship ofsialic acid to
function and in vivo survival of human factor Vill/von Willebrand
factor protein. I Biol Chem 252:5538-5546, 1977
65. Gralnick HR: Factor VIII/von Willebrand factor protein:
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
THE FACTOR VIII COMPLEX 13
Galactose, a cryptic determinant of von Willebrand factor activity. I
Clin Invest 62:496-499, 1978
66. Sodetz II, Paulson IC, Pizzo SV, et al: Carbohydrate on
human factor VIlI/von Willebrand factor: Impairment of function
by removal of specific galactose residues. I Biol Chem 253:7202-
7206, 1978
67. Hoyer LW, de los Santos R, Hoyer JR: Antihemophilic
factor antigen. Localization in endothelial cells by immunofluores-
cent microscopy. J Clin Invest 52:2737-2744, 1973
68. Bloom AL, Giddings IC, Wilks Ci: Factor VIII on the
vascular intima: Possible importance in haemostasis and thrombosis.
Nature (New Biol) 241:217-219, 1973
69. Jaffe EA, Hoyer LW, Nachman RL: Synthesis of anti-
hemophilic factor antigen by cultured human endothelial cells. I
Clin Invest 52:2757-2764, 1973
70. laffe EA, Hoyer LW, Nachman RL: Synthesis ofvon Wille-
brand factor by cultured human enodthelial cells. Proc NatI Acad
Sci USA 71:1906-1909, 1974
71. Tuddenham EGD, Lazarchick J, Hoyer LW: Synthesis and
release of factor VIII bycultured human endothelial cells. Br I
Haematol 47:617-626, 1981
72. Rizza CR, Rhymes IL, Austen DEG, Kernoff PBA, Aroni
SA: Detection of carriers of haemophilia: a “blind” study. Br J
Haematol 30:447-456, 1975
73. Bird P. Rizza CR: A method for detecting factor VIII
clotting activity associated with factor VIlI-related antigen in
agarose gels. Br I Haematol 31:5-12, 1975
74. Hoyer LW: Specificity of precipitating antibodies in immu-
nologic identification of antihaemophilic factor. Nature (New Biol)
245:49-51, 1973
75. Lazarchick I, I-Ioyer LW: The properties of immune
complexes formed by human antibodies to factor VIII. I Clin Invest
60:1070-1079, 1977
76. Blomb#{228}ck B, Hessel B, Savidge G, Wikstr#{246}m L, Blomb#{228}ck
M: The effect of reducing agents on factor VIII and other coagula-
tion factors. Throm Res 12:1 177-1 194, 1978
77. Weiss Hi, Sussman II, Hoyer LW: Stabilization of factor
VIII in plasma by the von Willebrand factor. I Clin Invest 60:390-
404, 1977
78. Denson KWE, Biggs R, Haddon ME, Borrett R, Cobb K:
Two types of haemophilia (A + and A - ): A study of 48 cases. Br I
Haematol 17:163-171, 1969
79. Feinstein D, Chong MNY, Kasper CK, Rapaport SI: Hemo-
philia A: Polymorphism detectable by a factor VIII antibody.
Science 163:1071-1072, 1969
80. Reisner HM, Price WA, Blatt PM, Barrow ES, Graham JB:
Factor VIII coagulant antigen in hemophilic plasma: A comparison
of five alloantibodies. Blood 56:6 1 5-6 19, 1980
8 1 . Hoyer LW, Shapiro 55: Unpublished studies
82. Zimmerman TS, Ratnoff OD, Littell AS: Detection of
carriers of classic hemophilia using an immunologic assay for
antihemophilic factor (factor VIII). I Clin Invest 50:255-258, 1971
83. Klein HG, Aledort LM, Bouma BN, Hoyer LW, Zimmer-
man TS, DeMets DL: A cooperative study for the detection of the
carrier state of classic hemophilia. N EngI I Med 296:959-962,
I977
84. Firshein SI, Hoyer LW. Lazarchick I, Forget BG, Hobbins
IC, Clyne LP, Pilick FA, Muir WA, Merkatz IR, Mahoney MI:
Prenatal diagnosis of classic haemophilia. N EngI I Med 300:937-
94l, 1979
85. Mibashan RS, Rodeck CH, Furlong RA, Bains L, Peake IR,
Thumpston 1K, Gorer R, Bloom AL: Dual diagnosis of prenatal
haemophilia by measurement of fetal factor VIIIC and VIIIC
antigen (VIIICAg). Lancet 2:994-997, 1980
86. Hoyer LW, Mahoney Mi: Prenatal diagnosis of classic
hemophilia. Proc. XIV Int. Congress. World Fed. Hemophilia, San
Jose, Costa Rica, July 3-7, 1981
87. Bloom AL: The von Willebrand syndrome. Semin Hematol
17:215-227, 1980
88. Abildgaard CF. Suzuki Z. Harrison I, iefcoat K, Zimmer-
man TS: Serial studies in von Willebrand’s disease: Variability
versus “variants.” Blood 56:712-716, 1980
89. Zimmerman TS. Abildgaard CF. Meyer D: The factor VIII
abnormality in severe von Willebrand’s disease. N EngI I Med
301:1307-1310, 1979
90. Kernoff PBA, Gruson R, Rizza CR: A variant of factor VIII
related antigen. Br I Haematol 26:435-440, 1974
91. Peake IR, Bloom AL. Giddings IC: Inherited variants of
factor VIll-related protein in von Willebrand’s disease. N EngI I
Med 291:1 13-1 17, 1974
92. Meyer D, Obert B, Pietu G, Lavergne IM, Zimmerman IS:
Multimeric structure of factor VIlI/von Willebrand factor in Von
Willebrand’s disease. I Lab Clin Med 95:590-602, 1980
93. Gralnick HR, Coller BS: Carbohydrate deficiency of the
factor VIII/von Willebrand factor protein in von Willebrand’s
disease variants. Science I 92:56-59, 1976
94. Gralnick HR, Sultan Y, Coller BS: von Willebrand’s disease.
Combined qualitative and quantiative abnormalities. N EngI I Med
296:1024-1030, 1977
95. Zimmerman IS, Voss R, Edgington IS: Carbohydrate of the
factor VIlI/von Willebrand factor in von Willebrand’s disease. I
Clin lnvest64:l298-1302, 1979
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
1981 58: 1-13
LW Hoyer The factor VIII complex: structure and function
http://www.bloodjournal.org/content/58/1/1.full.htmlUpdated information and services can be found at:
Articles on similar topics can be found in the following Blood collections
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:
Copyright 2011 by The American Society of Hematology; all rights reserved.Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of
For personal use only.on November 9, 2017. by guest www.bloodjournal.orgFrom
Recommended